Prosthetic alignment method and device

By extracting the features in the wearable image of the prosthetic limb and calculating the relative position change data, identifying abnormal gaits, and automatically determining the prosthetic limb line strategy, the problem of prosthetic limb line relies on human eye observation in the existing technology, and improving the adaptability of the line strategy Matching and comfort.

CN118430020BActive Publication Date: 2024-11-22NAT REHABILITATION ASSISTIVE DEVICES RES CENT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410599780.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-15
Publication Date
2024-11-22
Estimated Expiration
2044-05-15

AI Technical Summary

Technical Problem

In the prior art, prosthetic line relies on the naked eye observation and qualitative analysis of the prosthetician, making adjustments time-consuming and labor-intensive and difficult to ensure high fit.

Method used

By obtaining the prosthetic wear image sequence, prosthetic line characteristics are extracted, such as accepting cavity axis features, joint node features and plantar plane features, calculating motion trajectory and relative position change data, identifying abnormal gaits, and determining line strategies based on personal information .

Benefits of technology

It improves the rationality and adaptability of the prosthetic line strategy, reduces the influence of human factors, and ensures higher suitability and comfort.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118430020B_ABST
    Figure CN118430020B_ABST
Patent Text Reader

Abstract

The present invention provides a prosthesis alignment method and device, including: acquiring a prosthesis wearing image sequence, segmenting and obtaining a human body contour and a prosthesis contour in each prosthesis wearing image; extracting the prosthesis alignment features corresponding to each prosthesis wearing image based on each human body contour and prosthesis contour, the prosthesis alignment features including the receiving cavity axis feature, the joint point feature and the plantar plane feature; determining the prosthesis alignment feature motion trajectory based on each prosthesis alignment feature, determining the relative position change data of the prosthesis alignment feature based on the prosthesis alignment feature motion trajectory, the prosthesis alignment feature motion trajectory including the receiving cavity axis motion trajectory, the joint point motion trajectory and the plantar plane motion trajectory; identifying abnormal gait based on the prosthesis alignment feature motion trajectory and the relative position change data; acquiring the personal information of the prosthesis wearer, and determining the prosthesis alignment strategy based on the abnormal gait and the personal information. The present invention can provide a prosthesis alignment strategy with a high degree of adaptability for the prosthesis wearer.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of prostheses, and in particular to a prosthesis alignment method and device. Background Art

[0002] Prosthetic configuration refers to the general term for all professional activities in the whole process of installing and equipping prostheses for amputees and finally delivering finished prostheses, including patient assessment, prosthetic prescription, production and fitting, usage training, prosthetic function assessment, and prosthetic delivery.

[0003] Alignment is one of the cores of prosthesis making and prosthesis fitting procedures. It is to determine the relative position relationship between the prosthesis and the body as a whole and the ground. It is generally divided into three steps: workbench alignment, static alignment and dynamic alignment. During the alignment adjustment process, the geometric configuration of the prosthesis is changed to continuously correct the stress load mode between the residual limb and the ground, and an energy-efficient stride is created by reducing the lateral and longitudinal center of gravity offset. The goal of the optimal force line is stable support phase, clear swing phase, consistent step length and high energy efficiency; specifically, alignment adjustment also includes: 1. Transition from initial heel contact to full sole landing; 2. Stability of the knee joint and control of the pelvis during the support phase; 3. Smooth forward propulsion (rolling) during the transition from early to late support phase; 4. Ensure sufficient swing phase toe clearance.

[0004] The quality of prosthesis alignment greatly affects the comfort and functionality of the prosthesis. Unsatisfactory alignment can also cause various abnormal gaits. Therefore, it is necessary to achieve better alignment of the prosthesis during the prosthesis fitting and installation process. Prosthesis alignment in the prior art has always relied entirely on the prosthetist's visual observation and qualitative analysis for evaluation and adjustment. The evaluation results and adjustment plans rely on the prosthetist's skills and experience, which is not only time-consuming and labor-intensive, but also difficult to ensure a high degree of fit. Therefore, how to determine a prosthesis alignment strategy that has a high degree of fit with the prosthesis wearer is a technical problem that needs to be solved urgently. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a prosthesis alignment method and device to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a prosthesis alignment method, the method comprising the following steps:

[0007] Acquire a sequence of prosthesis wearing images, and segment and obtain the human body contour and prosthesis contour in each prosthesis wearing image;

[0008] Extracting prosthesis alignment features corresponding to each prosthesis wearing image based on the obtained human body contour and prosthesis contour in each prosthesis wearing image, wherein the prosthesis alignment features include a receiving cavity axis feature, a joint point feature, and a plantar plane feature;

[0009] Determine the prosthesis alignment feature motion trajectory based on the prosthesis alignment feature corresponding to each of the prosthesis wearing images, and determine the relative position change data of the prosthesis alignment feature based on the prosthesis alignment feature motion trajectory, wherein the prosthesis alignment feature motion trajectory includes the receiving cavity axis motion trajectory, the joint point motion trajectory and the plantar plane motion trajectory;

[0010] Identify abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data;

[0011] The personal information of the prosthesis wearer is obtained, and a prosthesis alignment strategy is determined based on the abnormal gait and the personal information.

[0012] In some embodiments of the present invention, extracting the prosthesis alignment features corresponding to each of the prosthesis wearing images based on the human body contour and the prosthesis contour in each of the prosthesis wearing images obtained includes:

[0013] Extracting a socket contour from a prosthesis contour, determining a human limb axis based on the human body contour, and obtaining a socket axis feature based on fitting the socket contour and the human limb axis;

[0014] Based on the human body contour and the prosthesis contour, joint point features are obtained through a trained joint point recognition model;

[0015] A foot contour is extracted from the human body contour and the prosthesis contour, and plantar plane features are determined based on the extracted foot contour.

[0016] In some embodiments of the present invention, the plantar plane feature includes foot center point data and a plantar normal vector; and / or,

[0017] The receiving cavity axis feature is obtained based on the receiving cavity contour and the human limb axis fitting, including:

[0018] Based on the socket contour and the human limb axis, a least square method or a principal component analysis method is used to preliminarily fit the socket axis;

[0019] Determine limb movement characteristics based on the human limb axis corresponding to each prosthesis wearing image in the prosthesis wearing image sequence;

[0020] The receiving cavity axis characteristics are obtained by optimizing the initially fitted receiving cavity axis based on the limb movement characteristics.

[0021] In some embodiments of the present invention, segmenting and obtaining the human body contour and the prosthesis contour in each prosthesis wearing image includes:

[0022] Removing noise from the prosthesis wearing image based on a filtering algorithm;

[0023] The prosthesis wearing images based on noise removal are segmented by threshold segmentation method, region growing method or trained contour segmentation model to obtain the human body contour and prosthesis contour in each prosthesis wearing image.

[0024] In some embodiments of the present invention, the relative position change data include: the motion trajectory change data of the receiving cavity axis, the motion trajectory change data of the joint point, the motion trajectory change data of the plantar plane, the shortest distance between the joint point and the receiving cavity axis, the distance between the joint point and the plantar plane, and the angle between the receiving cavity axis and the plantar plane.

[0025] In some embodiments of the present invention, identifying abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data includes: obtaining the abnormal gait through a trained abnormal gait recognition model based on the prosthesis alignment characteristic motion trajectory and the relative position change data.

[0026] In some embodiments of the present invention, determining a prosthesis alignment strategy based on the abnormal gait and the personal information includes: determining a prosthesis alignment strategy through a trained solution recommendation model based on the abnormal gait and the personal information.

[0027] In some embodiments of the present invention, the method further comprises:

[0028] constructing an initial abnormal gait recognition model and a first sample data set, and training the initial abnormal gait recognition model based on the first sample data set to obtain a trained abnormal gait recognition model; and / or,

[0029] Constructing an initial solution recommendation model and a second sample data set, and training the initial solution recommendation model based on the second sample data set to obtain a trained solution recommendation model;

[0030] Among them, the sample data in the first sample data set includes prosthesis alignment feature motion trajectory sample data, relative position change sample data and abnormal gait sample data, and the sample data in the second sample data set includes abnormal gait sample data, personal information sample data and prosthesis alignment strategy sample data.

[0031] According to another aspect of the present invention, a prosthetic alignment system is disclosed, including a processor, a memory, and a computer program stored in the memory, wherein the processor is used to execute the computer program, and when the computer program is executed, the system implements the steps of the method described in any of the above embodiments.

[0032] According to yet another aspect of the present invention, a computer-readable storage medium is disclosed, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0033] The prosthesis alignment method and device of the present invention extracts prosthesis alignment features from each prosthesis wearing image when determining the prosthesis alignment strategy, and the prosthesis alignment features include the axis features of the receiving cavity, the features of the joint points, and the features of the plantar plane. In addition, the motion trajectory of the prosthesis alignment features and the relative position change data of the prosthesis alignment features are determined, and the abnormal gait is identified based on the determined prosthesis alignment feature motion trajectory and relative position change data, and finally the final prosthesis alignment strategy is determined based on the wearer's personal information and abnormal gait. This method not only considers the prosthesis alignment features when determining the prosthesis alignment strategy, but also comprehensively considers the motion trajectory and relative position change data of the prosthesis alignment features, thereby ensuring the rationality of the determined prosthesis alignment strategy, and further improving the high adaptability of the prosthesis alignment strategy to the prosthesis wearer.

[0034] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0035] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of the present application, and do not constitute a limitation of the present invention. In the drawings:

[0037] Figure 1 Schematic diagram of the flow of a prosthesis alignment method in one embodiment of the present invention.

[0038] Figure 2 FIG. 4 is a flow chart of a prosthesis alignment method according to another embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0040] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0041] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0042] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0043] The prosthesis alignment method of the present application applies machine vision technology and expert system to automatically obtain the prosthesis alignment strategy by capturing the human body contour and prosthesis contour, extracting prosthesis alignment features, calculating the prosthesis alignment feature motion trajectory and relative position change data, identifying abnormal gait patterns, and recommending abnormal gait alignment adjustment plans. The method is simple and easy to use, and is not affected by the professionalism and experience of the prosthetist when determining the prosthesis alignment strategy.

[0044] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0045] Figure 1 FIG. 1 is a flow chart of a prosthesis alignment method according to an embodiment of the present invention. Figure 1 As shown, the prosthesis alignment method at least includes steps S10 to S50.

[0046] Step S10: Acquire a sequence of prosthesis wearing images, and segment and obtain the human body contour and prosthesis contour in each prosthesis wearing image.

[0047] In this step, the prosthesis wearing image sequence may be a plurality of prosthesis wearing images corresponding to different times, and the prosthesis wearing images include actions performed by the prosthesis wearer wearing the prosthesis. For example, in one embodiment, when the prosthesis wearer wears the prosthesis to run and walk, the prosthesis wearing image sequence is a series of prosthesis wearing images of the prosthesis wearer wearing the prosthesis to walk. In the prosthesis wearing image sequence, the two prosthesis wearing images before and after may have temporal continuity.

[0048] Exemplarily, in this step, a depth camera or a stereo camera system may be used to capture a three-dimensional real-time image, and multiple images of wearing prostheses in the three-dimensional real-time image are combined into a series of images that change over time. After capturing the prosthesis image sequence, the human body and prosthesis in each prosthesis wearing image are further segmented to obtain the human body contour and prosthesis contour in each prosthesis wearing image. That is, in this embodiment, based on the three-dimensional real-time image and the image recognition algorithm, the human body and prosthesis contour features of the prosthesis wearer are obtained, so that in the subsequent steps, based on the extracted human body contour and prosthesis contour, the prosthesis alignment features can be extracted and optimized in combination with the differences in the movement of each part.

[0049] In one embodiment, segmenting to obtain the human body contour and the prosthesis contour in each prosthesis wearing image may specifically include: removing noise in the prosthesis wearing image based on a filtering algorithm; segmenting the prosthesis wearing image from which the noise has been removed by a threshold segmentation method, a region growing method or a trained contour segmentation model to obtain the human body contour and the prosthesis contour in each prosthesis wearing image.

[0050] In the above embodiment, after the noise in the image is removed by a filtering algorithm, the contours of the human body and the prosthesis are segmented using threshold segmentation, region growing or deep learning algorithms, wherein the filtering algorithm may be a median filter or a Gaussian filter. In addition, in the process of segmenting the human body contour and the prosthesis contour, an edge detection algorithm (such as Canny edge detection) may be applied to identify the edge of the contour, and a polygon fitting or curve fitting method may be used to connect the edge points into a smooth contour.

[0051] Among them, the input of the contour segmentation model is the prosthesis wearing image with noise removed, and the output is the segmented human body contour and prosthesis contour. In order to obtain a trained contour segmentation model, the initial contour segmentation model, the sample data set and the model loss function can be constructed first, and then the initial contour segmentation model is pre-trained based on the sample data set and the model loss function to finally obtain the trained contour segmentation model. It can be understood that the sample data in the sample data set used to iteratively update the parameters of the initial contour segmentation model includes the prosthesis wearing image sample data and the label features corresponding to the prosthesis wearing image sample data, and the label features corresponding to the prosthesis wearing image sample data include body contour label information and prosthesis contour label information.

[0052] Step S20: extracting the prosthesis alignment features corresponding to each prosthesis wearing image based on the human body contour and prosthesis contour in each prosthesis wearing image, wherein the prosthesis alignment features include the receiving cavity axis feature, the joint point feature and the plantar plane feature.

[0053] This step is to further extract the prosthesis alignment features corresponding to each prosthesis wearing image from the segmented human body contour and prosthesis contour. The type of joint point is specifically determined according to the amputation site of the prosthesis wearer. For example, when the prosthesis wearer has a thigh amputation, the joint points include the knee joint point and the ankle joint point, and the joint point features include the knee joint point features and the ankle joint point features; and when the prosthesis wearer has a calf amputation, the joint points only include the ankle joint point, and the joint point features are ankle joint point features.

[0054] Exemplarily, extracting the prosthesis alignment features corresponding to each prosthesis wearing image based on the human body contour and prosthesis contour obtained in each prosthesis wearing image may specifically include the following steps: extracting the socket contour from the prosthesis contour, determining the human limb axis based on the human body contour, and obtaining the socket axis feature based on the fitting of the socket contour and the human limb axis; obtaining the joint feature based on the human body contour and the prosthesis contour through a trained joint recognition model; extracting the foot contour from the human body contour and the prosthesis contour, and determining the plantar plane feature based on the extracted foot contour.

[0055] Furthermore, the characteristics of the receiving cavity axis are obtained based on the fitting of the receiving cavity contour and the human limb axis, including: preliminarily fitting the receiving cavity axis through the least squares method or the principal component analysis method based on the receiving cavity contour and the human limb axis; determining the limb movement characteristics based on the human limb axis corresponding to each prosthesis wearing image in the prosthesis wearing image sequence; and obtaining the characteristics of the receiving cavity axis by optimizing the preliminarily fitted receiving cavity axis based on the limb movement characteristics.

[0056] Among them, the receiving cavity axis feature can be specifically the receiving cavity axis equation, the joint point feature can be specifically the position coordinates of the joint point, and the plantar plane feature specifically includes the foot center point data and the plantar normal vector, and the foot center point data is the position coordinates of the foot center point.

[0057] In some embodiments, after the prosthesis alignment feature is extracted, it can be further optimized to obtain the final prosthesis alignment feature. Exemplarily, the prosthesis alignment feature extraction step includes the steps of extracting and fitting the socket axis equation, detecting and optimizing the joint feature, and extracting the contour of the foot plane and the plane fitting step. In the step of extracting and fitting the socket axis equation, the socket position is determined in the prosthesis contour according to the prosthesis structure and shape, and the socket contour is segmented from the entire contour; the least squares method or the statistical method of principal component analysis is used to preliminarily fit the socket axis according to the human limb axis and the socket shape, and the fitting result is optimized according to the limb movement characteristics to determine the optimized socket axis equation. In the joint feature detection and position optimization step, the joint points are identified by shape analysis or deep learning algorithm; the detected joint point position is fine-tuned by optimization algorithm (such as least squares method) according to the characteristics of each part during movement (such as range of motion, movement trajectory) to make it more accurate; in one embodiment, the joint points include knee joint points and ankle joint points, so in this step, the knee joint points and ankle joint points are identified by shape analysis or deep learning algorithm. In the steps of contour extraction and plane fitting of the foot plane, the foot portion is separated from the entire contour, and the equation of the plantar plane is determined by fitting the foot contour points.

[0058] Exemplarily, the prosthesis alignment feature is defined as follows: using a three-dimensional coordinate vector J = (Jx , J y , J z ) represents the position coordinates of the joint point J; through two points S1 = (S 1x , S 1y , S 1z ) and S2=(S 2x , S 2y , S 2z ) defines the axis of the receiving cavity, the direction vector of the axis of the receiving cavity is A = S1-S2, and the unit direction vector is The foot center point data is expressed as F C =(F Cx , F Cy , F Cz ), the plantar normal vector is represented by N F =(N Fx , N Fy , N Fz ).

[0059] Step S30: Determine the prosthesis alignment feature motion trajectory based on the prosthesis alignment features corresponding to each of the prosthesis wearing images, and determine the relative position change data of the prosthesis alignment features based on the prosthesis alignment feature motion trajectory, wherein the prosthesis alignment feature motion trajectory includes the receiving cavity axis motion trajectory, the joint point motion trajectory and the plantar plane motion trajectory.

[0060] In this step, based on the extracted prosthesis alignment features and the three-dimensional real-time image sequence when the prosthesis wearer walks, the changes in the motion trajectory and relative position relationship of the prosthesis alignment features are calculated. The relative position change data may include: the motion trajectory change data of the receiving cavity axis, the motion trajectory change data of the joint point, the motion trajectory change data of the plantar plane, the shortest distance between the joint point and the receiving cavity axis, the distance between the joint point and the plantar plane, and the angle between the receiving cavity axis and the plantar plane. That is, when calculating the relative position change data between the prosthesis alignment features, the prosthesis alignment features can be tracked in real time in the three-dimensional real-time image sequence, and the spatial coordinate difference, vector angle and trajectory difference between the features can be calculated to obtain the distance, relative rotation and torsion between the features, and then obtain the change pattern of the relative position relationship between the features during the entire movement process.

[0061] Furthermore, taking the above-defined prosthesis alignment features as an example, the motion trajectory of the prosthesis alignment features of the present application and the method for calculating the relative position change data between the prosthesis alignment features are described in detail.

[0062] Joint point J = (J x , J y , J z)’s motion trajectory in three-dimensional space is represented by a series of coordinate points that change with time. The coordinates of the joint point J on the x, y, and z axes at time t are J(t) = (Jx(t), Jy(t), Jz(t)); the motion trajectory of the joint point J is described as a set of coordinate points that change with time, that is, the joint point motion trajectory = {J(t)|t∈time domain}; wherein the time domain can be continuous (for example, t∈[0,+∞)) or discrete (for example, a series of specific time points).

[0063] The receiving cavity axis passes through two points S1 = (S 1x , S 1y , S 1z ) and S2=(S 2x , S 2y , S 2z ), then the motion trajectory of the receiving cavity axis is represented by the motion trajectory of the two points S1 and S2. Specifically, the coordinates of S1 and S2 corresponding to time t are S1(t)=(S 1x (t), S 1y (t), S 1z (t)) and S2(t)=(S 2x (t), S 2y (t), S 2z (t)), the corresponding motion trajectory of the receiving cavity axis = {{S1(t), S2(t)}|t∈time domain}.

[0064] The plantar plane features include the foot center point data F C =(F Cx , F Cy , F Cz ) and the plantar normal vector data N F =(N Fx , N Fy , N Fz ), that is, the motion trajectory of the plantar plane feature is represented by the motion trajectory of the foot center point and the change of the normal vector. The coordinates of the foot center point at time t are F C (t)=(F Cx (t), F Cy (t), F Cz (t)), the normal vector is N F (t)=(N Fx (t), N Fy (t), N Fz (t)), then the corresponding plantar motion trajectory = {{F C (t), N F (t)}|t∈time domain}.

[0065] After obtaining the motion trajectory of the prosthesis alignment feature, the motion trajectory change data of the prosthesis alignment feature is further calculated, that is, the motion trajectory change data of the axis of the receiving cavity is calculated based on the motion trajectory of the receiving cavity axis, the motion trajectory change data of the joint point is calculated based on the motion trajectory of the joint point, and the motion trajectory change data of the plantar plane is calculated based on the motion trajectory of the plantar plane. Taking joint point J as an example, J x (t), J y (t), J z (t) is the function of the joint point J changing with time t on the x, y, and z coordinate axes. The velocity vector of the joint point can be expressed as the derivative of the position vector with respect to time. Then the velocity vector The acceleration vector of joint point J can be expressed as the derivative of the velocity vector with respect to time, so the acceleration vector

[0066] Similarly, the velocity of the axis of the receiving cavity during the movement is obtained by differentiating the above S1(t) and S2(t), and the acceleration of the axis of the receiving cavity during the movement is obtained by differentiating the velocity of the axis of the receiving cavity during the movement. The velocity of the plantar plane during the movement is obtained by differentiating the above foot center point F C (t) and the normal vector N F (t) is derived, and the acceleration of the plantar plane during the movement is obtained by differentiating the velocity of the plantar plane during the movement.

[0067] In addition, the trajectory change data of the prosthesis alignment feature also includes the curvature of the trajectory, the torsion of the trajectory, the length of the trajectory, and the rate of change of acceleration (jerk, also known as jump). For example, after calculating the velocity vector and acceleration vector of the joint point, the curvature of the joint point motion trajectory can be further calculated. The curvature of the joint point motion trajectory indicates how fast the direction of the joint point changes under the unit speed on the trajectory. The torsion of the trajectory indicates the rotation rate of the trajectory in the tangential direction. where b J (t) is the binormal vector of the joint point’s motion trajectory, The length of the joint point motion trajectory The rate of change of acceleration of the joint point is the jerk, |j(t)| reflects the smoothness and comfort of movement.

[0068] In addition to the above, the relative position change data also includes relative position relationship information, such as the shortest distance between the joint point and the receiving cavity axis, the distance between the joint point and the plantar plane, and the angle between the receiving cavity axis and the plantar plane. In one embodiment, the shortest distance from the joint point J to the receiving cavity axis is calculated using the point-to-straight line distance formula. Among them, × represents the cross product operation of the vector, and || represents the modulus (length) operation of the vector. Use the point-to-plane distance formula to calculate the distance from the joint point J to the plantar plane. Where · represents the dot product operation of the vector. The angle between the axis of the socket and the plantar plane is calculated using the formula for the vector angle.

[0069] Step S40: identifying abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data.

[0070] In this step, based on the acquired prosthetic alignment feature motion trajectory and the change of relative position relationship, the abnormal gait is identified. Exemplarily, the identification of abnormal gait can be achieved by a pattern recognition algorithm, which can be implemented based on a decision tree classifier.

[0071] In a specific embodiment, identifying abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data includes: obtaining abnormal gait through a trained abnormal gait recognition model based on the prosthesis alignment characteristic motion trajectory and the relative position change data. In this embodiment, the input of the abnormal gait recognition model is the prosthesis alignment characteristic motion trajectory and the relative position change data, and the output is the abnormal gait.

[0072] In order to obtain a trained abnormal gait recognition model, the prosthesis alignment method may further include the following steps: constructing an initial abnormal gait recognition model and a first sample data set, and training the initial abnormal gait recognition model based on the first sample data set to obtain a trained abnormal gait recognition model. The sample data in the first sample data set includes prosthesis alignment feature motion trajectory sample data, relative position change sample data, and abnormal gait sample data.

[0073] Step S50: Obtaining personal information of the prosthesis wearer, and determining a prosthesis alignment strategy based on the abnormal gait and the personal information.

[0074] After the abnormal gait is identified, a prosthesis alignment strategy that is highly matched with the abnormal gait is further determined based on the identified abnormal gait and the specific personal information of the prosthesis wearer. When determining the prosthesis alignment strategy, the cause of the abnormal gait can be determined based on the identified abnormal gait using a conditional formula or a rule-based classifier, and then based on the cause, a solution is queried and recommended through a pre-made list to reduce gait deviation. Exemplarily, the personal information of the prosthesis wearer may include medical history, strength, endurance, activity level, complications, socket fit, prosthesis components, previous experience in wearing prostheses, residual limb volume changes, and shoe shape features. Therefore, in this embodiment, when using a conditional formula or a rule-based classifier to determine the cause of the abnormal gait, in order to improve the accuracy of the determined cause of the abnormal gait, the personal information of the prosthesis wearer is also taken into account.

[0075] In one embodiment, determining the prosthesis alignment strategy based on the abnormal gait and the personal information specifically includes: determining the prosthesis alignment strategy through a trained solution recommendation model based on the abnormal gait and the personal information. In this case, the input features of the solution recommendation model are the abnormal gait and the personal information, and the output of the solution recommendation model is the prosthesis alignment strategy.

[0076] Furthermore, in order to obtain a trained solution recommendation model, the prosthesis alignment method further includes the following steps: constructing an initial solution recommendation model and a second sample data set, and training the initial solution recommendation model based on the second sample data set to obtain a trained solution recommendation model. It can be understood that at this time, the sample data in the second sample data set includes abnormal gait sample data, personal information sample data, and prosthesis alignment strategy sample data.

[0077] When the prosthesis alignment method disclosed in the above-mentioned embodiment is used to generate a prosthesis alignment strategy, when the gait pattern of the prosthesis wearer is finally detected to be safe, comfortable, smooth, stable, efficient and energy-saving, it means that there is no abnormal gait. If both the prosthesis wearer and the prosthetist can accept the prosthesis alignment strategy at this time, the final alignment of the prosthesis can be determined based on the prosthesis alignment strategy; otherwise, the above-mentioned steps of the prosthesis alignment method disclosed in the embodiment of the present invention can be repeated until the gait pattern of the prosthesis wearer has no abnormal gait.

[0078] In another embodiment, if Figure 2 As shown, the prosthesis alignment method of the present application applies machine vision technology and expert system to automatically obtain the prosthesis alignment strategy by capturing the human body contour and prosthesis contour, extracting prosthesis alignment features, calculating the prosthesis alignment feature motion trajectory and relative position change data, identifying abnormal gait patterns, and recommending abnormal gait alignment adjustment plans.

[0079] Correspondingly, the present invention also provides a prosthetic alignment system, comprising a processor, a memory and a computer program stored in the memory, wherein the processor is used to execute the computer program, and when the computer program is executed, the system implements the steps of the method described in any of the above embodiments.

[0080] Through the above embodiments, it can be found that the prosthesis alignment method and system of the present invention extracts prosthesis alignment features from each prosthesis wearing image when determining the prosthesis alignment strategy, and the prosthesis alignment features include the axis features of the receiving cavity, the joint point features and the plantar plane features. In addition, the motion trajectory of the prosthesis alignment features and the relative position change data of the prosthesis alignment features are determined, and the abnormal gait is identified based on the determined prosthesis alignment feature motion trajectory and relative position change data, and finally the final prosthesis alignment strategy is determined based on the wearer's personal information and abnormal gait. This method not only considers the prosthesis alignment features when determining the prosthesis alignment strategy, but also comprehensively considers the motion trajectory and relative position change data of the prosthesis alignment features, thereby ensuring the rationality of the determined prosthesis alignment strategy, and thus also improving the high adaptability of the prosthesis alignment strategy to the prosthesis wearer.

[0081] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned edge computing server deployment method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0082] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0083] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0084] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0085] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A prosthesis alignment method, characterized in that: The method comprises the following steps: Acquire a sequence of prosthesis wearing images, and segment and obtain the human body contour and prosthesis contour in each prosthesis wearing image; Extracting prosthesis alignment features corresponding to each prosthesis wearing image based on the obtained human body contour and prosthesis contour in each prosthesis wearing image, wherein the prosthesis alignment features include a receiving cavity axis feature, a joint point feature, and a plantar plane feature; Determine the prosthesis alignment feature motion trajectory based on the prosthesis alignment feature corresponding to each of the prosthesis wearing images, and determine the relative position change data of the prosthesis alignment feature based on the prosthesis alignment feature motion trajectory, wherein the prosthesis alignment feature motion trajectory includes the receiving cavity axis motion trajectory, the joint point motion trajectory and the plantar plane motion trajectory; Identify abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data; Obtain personal information of the prosthesis wearer, and determine a prosthesis alignment strategy based on the abnormal gait and the personal information; wherein the personal information includes medical history, strength, endurance, activity level, complications, socket fit, prosthetic components, previous experience with wearing prostheses, residual limb volume changes, and shoe shape characteristics.

2. The prosthesis alignment method according to claim 1, characterized in that: Extracting the prosthesis alignment features corresponding to each prosthesis wearing image based on the human body contour and the prosthesis contour in each prosthesis wearing image obtained, including: Extracting a socket contour from a prosthesis contour, determining a human limb axis based on the human body contour, and obtaining a socket axis feature based on fitting the socket contour and the human limb axis; Based on the human body contour and the prosthesis contour, joint point features are obtained through a trained joint point recognition model; A foot contour is extracted from the human body contour and the prosthesis contour, and plantar plane features are determined based on the extracted foot contour.

3. The prosthesis alignment method according to claim 2, characterized in that: The plantar plane features include the foot center point data and the plantar normal vector; and / or, The receiving cavity axis feature is obtained based on the receiving cavity contour and the human limb axis fitting, including: Based on the socket contour and the human limb axis, a least square method or a principal component analysis method is used to preliminarily fit the socket axis; Determine limb movement characteristics based on the human limb axis corresponding to each prosthesis wearing image in the prosthesis wearing image sequence; The receiving cavity axis characteristics are obtained by optimizing the initially fitted receiving cavity axis based on the limb movement characteristics.

4. The prosthesis alignment method according to claim 1, characterized in that: The human body contour and prosthesis contour in each prosthesis wearing image are obtained by segmentation, including: Removing noise from the prosthesis wearing image based on a filtering algorithm; The prosthesis wearing images based on noise removal are segmented by threshold segmentation method, region growing method or trained contour segmentation model to obtain the human body contour and prosthesis contour in each prosthesis wearing image.

5. The prosthesis alignment method according to claim 1, characterized in that: The relative position change data include: the motion trajectory change data of the receiving cavity axis, the motion trajectory change data of the joint point, the motion trajectory change data of the plantar plane, the shortest distance between the joint point and the receiving cavity axis, the distance between the joint point and the plantar plane, and the angle between the receiving cavity axis and the plantar plane.

6. The prosthesis alignment method according to claim 5, characterized in that: Identifying abnormal gait based on the prosthesis alignment characteristic motion trajectory and the relative position change data includes: obtaining abnormal gait through a trained abnormal gait recognition model based on the prosthesis alignment characteristic motion trajectory and the relative position change data.

7. The prosthesis alignment method according to claim 6, characterized in that: Determining a prosthesis alignment strategy based on the abnormal gait and the personal information includes: determining a prosthesis alignment strategy through a trained solution recommendation model based on the abnormal gait and the personal information.

8. The prosthesis alignment method according to claim 7, characterized in that: The method further comprises: constructing an initial abnormal gait recognition model and a first sample data set, and training the initial abnormal gait recognition model based on the first sample data set to obtain a trained abnormal gait recognition model; and / or, Constructing an initial solution recommendation model and a second sample data set, and training the initial solution recommendation model based on the second sample data set to obtain a trained solution recommendation model; Among them, the sample data in the first sample data set includes prosthesis alignment feature motion trajectory sample data, relative position change sample data and abnormal gait sample data, and the sample data in the second sample data set includes abnormal gait sample data, personal information sample data and prosthesis alignment strategy sample data.

9. A prosthetic alignment system, comprising a processor, a memory, and a computer program stored in the memory, characterized in that: The processor is used to execute the computer program. When the computer program is executed, the system implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.